Critical Appraisal of the SAHaRA Trial: Liberal versus Restrictive Red-Cell Transfusion Strategy in Aneurysmal Subarachnoid Hemorrhage

Aneurysmal subarachnoid hemorrhage (aSAH) is associated with high morbidity and mortality. Acute anemia, occurring in up to 50% of patients, has been linked to secondary brain injury, such as delayed cerebral ischemia (DCI). Red-cell transfusion, however, is also potentially dangerous, and the threshold at which benefit outweighs harm remains undefined. Whereas previous milestone trials like TRICC reported evidence that a restrictive transfusion strategy is safe among most critically ill patients,[ 1 ] it is uncertain whether it holds for the aSAH population. The SAHaRA (subarachnoid hemorrhage red-cell transfusion and outcome) trial sought to generate high-quality evidence to guide transfusion thresholds in this specific neurological setting.[ 2 ] The SAHaRA trial was a multicenter, randomized, open-label trial with blinded outcome assessment. Adult patients with aSAH and anemia (Hb ≤10 g/dL) within 10 days of admission were randomized to a liberal transfusion strategy (target Hb >10 g/dL) versus a restrictive strategy (target Hb >8 g/dL). The trial, conducted at 23 neurocritical care centers in the United States, Australia, and Canada, randomized 742 patients. The primary outcome was the proportion of patients with a modified Rankin Scale (mRS) score of ≥4 at 12 months, representing poor functional outcome. Secondary outcomes included quality-of-life measures, death, vasospasm, and delayed cerebral infarction. Randomization was computer-generated, center-stratified, and employed permuted blocks. While the trial was open-label, the blinding of outcome assessors strengthened internal validity. The two cohorts were well-matched at baseline for age (median 59 years), sex (82% female), World Federation of Neurosurgical Societies grade, and aneurysm location. Regarding the primary outcome, an mRS ≥4 was observed in 33.5% of the liberal cohort versus 37.7% of the restrictive cohort (risk ratio 0.88; 95% CI 0.72–1.09; p = 0.22), showing no statistically significant difference. On the other hand, regarding secondary outcomes, there were no differences in Functional Independence Measure scores, EuroQol 5-Dimension 5-Level questionnaire, Visual Analog Scale scores, or 12-month mortality. Notably, the liberal group had a lower rate of radiographic vasospasm (31.5 vs. 40.7%; RR 0.77; 95% CI 0.63–0.95), indicative of a probable physiological benefit. For safety outcomes, there was no increase in transfusion reactions, thromboembolic events, ARDS, or infection with the liberal approach. The trial possesses several strengths, including its robust design (randomization, stratification, and blinded assessors), strict protocol compliance, and international multicenter reach, which enhance external validity. However, some limitations warrant consideration. First, the trial may have been underpowered; it was designed to detect a 25% relative difference in primary outcome—an ambitious target. A smaller but clinically important difference might have been missed. Second, the open-label design allowed the treatment team to know the allocation, potentially affecting co-interventions. Third, a high rate of refusal (378 of 1,770 eligible patients) introduces potential selection bias. Finally, a Bayesian analysis might have yielded a more nuanced interpretation of the clinically important signal despite the lack of frequentist statistical significance. While the primary outcome was not significantly different, the direction of effect and the decrease in vasospasm indicate a signal of benefit for the liberal strategy without added harm. Such findings, in conjunction with the Transfusion Requirements in Acute Neurological Injury (TRAIN)[ 3 ] and Hemoglobin Transfusion Threshold in Traumatic Brain Injury Optimization[ 4 ] trials, support a transfusion threshold of Hb >9 g/dL within this population, particularly when the risk of vasospasm is increased. The SAHaRA trial provides valuable information on the ongoing controversy regarding transfusion in acute brain injury. In the absence of proven definitive superiority, the results indicate possible benefit without additional risk, supporting the argument for personalized, physiology-based management. A meta-analysis of individual patient data combining SAHaRA and TRAIN may provide further clarity. Until then, clinicians managing aSAH can reasonably target Hb levels >9 g/dL for patients at high risk of DCI or vasospasm. Publication History Article published online: 13 May 2026 © 2026. The Author(s). This is an open access article published by Thieme under the terms of the Creative Commons Attribution License, permitting unrestricted use, distribution, and reproduction so long as the original work is properly cited. (https://creativecommons.org/licenses/by/4.0/) Thieme Medical and Scientific Publishers Private Limited A-13A, Graphix Tower 1, 6th floor, Sector 62, Noida 201309, Uttar Pradesh, India